Linking Knowledge to Care: Knowledge Graph-Augmented Medical Follow-Up Question Generation

TL;DR

This paper presents KG-Followup, a knowledge graph-augmented language model that improves medical follow-up question generation by integrating domain knowledge, leading to better recall in pre-diagnostic assessments.

Contribution

Introduction of KG-Followup, a novel method combining knowledge graphs with LLMs for medical question generation, enhancing domain-specific reasoning capabilities.

Findings

Outperforms state-of-the-art methods by 5-8% in recall on relevant benchmarks.

Effectively integrates structured medical knowledge with LLMs for improved question relevance.

Demonstrates the importance of domain knowledge in medical NLP tasks.

Abstract

Clinical diagnosis is time-consuming, requiring intensive interactions between patients and medical professionals. While large language models (LLMs) could ease the pre-diagnostic workload, their limited domain knowledge hinders effective medical question generation. We introduce a Knowledge Graph-augmented LLM with active in-context learning to generate relevant and important follow-up questions, KG-Followup, serving as a critical module for the pre-diagnostic assessment. The structured medical domain knowledge graph serves as a seamless patch-up to provide professional domain expertise upon which the LLM can reason. Experiments demonstrate that KG-Followup outperforms state-of-the-art methods by 5% - 8% on relevant benchmarks in recall.